I listened with considerable interest to a BBC radio debate today relating to a published report claiming that 1.8million UK residents are "diagnosed" as suffering from CKD. http://www.bbc.co.uk/iplayer/episode/b01lt2vn/Inside_Health_Overdiagnosis/
On the program, Dr Donal O'Donaghue (Salford, UK) was defending such a growing vue calling for more detection and prevention as well as resources to manage this growing healthcare challenge. In effect, he was championing the prevaling, fashionable, view of rising CKD prevalence, the value of early detection and prevention and so many more often repeated but unsubstantiated and unproven assertions.
Dr Chritopher Winearls (Oxford, UK) was challenging this dogmatic vue, pointing out the inaccuracies of such prevalence statements, issues with detection and prevention programs as well as the grave concern over medicalisation of age-related decline in kidney function.
As often is the case, when the advocate of rising CKD are in trouble with their argument, they fall back on the "But CKD is a major CVD risk factor, so we need to detect it and prevent it...." argumement. One by the way totally refuted by most cardiologists and many nephrologists such as myself for the following reasons:
1. CKD is a MANIFESTATION of underlying CVD not its primary CAUSE, as I stated on many occassions on OLA; I even called the condition Cardio-Kidney-damage (C-K-D) in a 2010 Publication in KI.
2. In support of such argument is that subclinical CVD often precedes incident CKD in older individuals and accelerates its decline.
Here I refer to a number of publications showing that baseline atherosclerosis deteceted by raised carotid intima-media thickness predicts incident CKD (Choncol et al. http://www.ncbi.nlm.nih.gov/pubmed/18388124) and that those with subclinical cerebrovascular disease also have accelerated CKD decline within the community. http://www.ncbi.nlm.nih.gov/pubmed/20537454
This month in cJASN an article by Park et al. report on an observation from the Multi-Ethnic Study of Atherosclerosis (MESA) were they followed up 3,866 individuals (mean age 60) in the USA for approximately 5 years. They reported that those who go on to develop CKD have underlying CVD defined as subclinical cardiac hypertrophy detected by MRI. The presence of ventricular concentricity/hypertrophy increased risk of CKD develkopment (eGFR<60) but also that of increased risk of faster eGFR decline. This was independent of confounders such as hypertension and diabetes.
This supports once more the notion that older individuals with CKD have underlying CVD rather than the other way round. Consequently, it is not surprising that patients with CKD, a marker of underlying diffuse vascular pathology, have a worse CV outcome.
3. And perhaps more importantly, it is most likely that the underlying subclinical CVD that precede CKD may be the reflection of subclinical or underdiagnosed hypertension. In this week's Lancet a series of article stress the fact that systemic hypertension is both underdiagnosed and undertreated in developed as well as developing countries. The fact that casual/office blood pressure measurement is most unreliable and does not reflect 24h ambulatory blood pressure measurements or nocturnal hypertension known to be more closely associated with CVD.
It is likely that it is hypertension and the associated CVD that lead to CKD not the other way round.
30-40% of the population sufferes from hypertension.
Prevention and early Detection of Hypertension: YES, this should be the nephrologists' message!
Ultimately, CKD would, undoubtedly through hypertension and other uremia related factors, accelerate the progression of CVD; a vicious cycle described by Sir Richard Bright almost two centuries ago...!!!!
One of my favourite non-nephrology medical writers has to be Des Spence whose blog (http://bad4umedicine.blogspot.co.uk/) is rapidly becoming a must read. After previously dismissing chronic kidney disease as ' a failed experiment, a mockery of evidence-based medicine', Spence turns his attention to the $3 billion fine imposed Glaxo Smith Kline over amongst other things suppression of data concerning the cardiovascular risks of rosiglitazone. Spence points out that despite its' faults, pharma is essentially an industry ' doing untold good' and that its far too easy to blame pharma - he argues the real problem is a 'global medical culture' that encourages naivety and deference rather than respect and scepticism. In particular he is scathing about expert panels who he describes as 'educational mercenaries'.
Now whilst I don't agree all of his arguments reading his article reminded me of the time I visited a pharma-sponsored symposium at the WCN/ISN in Milan a few years ago. At that symposium Dr Geoffrey Block from Denver gave an impassioned talk on the evils of phosphorus in CKD. Phosphorus increased vascular calcification in CKD and the only way to reduce phosphorus was to use more dialysis and/or more phosphate binders (presumably the rather expensive non-calcium binders). What i remember more than anything was the call for phosphorus to be recognised as a cardiac toxin in CKD patients with a plea for food labelling - so that high phosphorus foods would receive some kind 'health warning' analogus to say high salt warnings that appear on much food packaging now. There was not much discussion of the negative DCOR study (a huge study showing sevelamer had no impact on mortality in dialysis patients) but instead a focus on how we probably needed to reduce phosphorus in the predialysis population.
So hows that all going? Well a recent publication in JASN online by Block suggests that the ' calcium bad, sevelamer/lanthanum good' narrative that has been pushed very hard by pharma (with I am afraid the collusion of some 'experts' in the CKD-MBD field), does not as yet stand up to any critical scrutiny...In this study 148 patients with estimated GFR=20–45 ml/min per 1.73 m2 were assigned to calcium acetate, lanthanum carbonate, sevelamer carbonate, or placebo. The primary endpoint was change in mean serum phosphorus from baseline whilst secondary endpoints looked at effects on FGF23, vascular calcification and bone mineral density. No surprise that the study met its primary endpoint - yes phosphate binders reduce serum phosphorus more effectively than placebo! As expected iPTH levels were significantly higher in the placebo group compared to the binder group at 9 months though interestingly FGF-23 levels were unchanged. In a subset of 90 patients (60 with binders and 26 placebo) imaging was performed which revealed some interesting results. Patients on phosphate binders had a small but significant increase in bone mineral density but a also a significant increase in progression of coronary artery calcification! The authors rather optimistically state that 'The apparent adverse effects on vascular calcification and salutary effects on BMD were most pronounced among patients randomized to calcium acetate' yet no statistics are provided - probably because only 20 patients on binders had an increase in calcification over 9 months and so any meaningful analysis of these 20 patients on three different binders was not possible!!
So what do I make of it all?
i) There is absolutely no RCT evidence to show that non-calcium binders are superior to calcium-based binders in the dialysis population on key patient centred outcomes such as death
ii) Whilst many have argued that the thresholds for serum phosphorus in the pre-dialysis population should be lowered, there is as yet no evidence to support this. In fact what this study shows is that further lowering of serum phosphorus with binders in predialysis patients with CKD may actually not be safe. Until an RCT comes out the honest answer is that we just dont know what to do..
iii) As nephrologists we can spend money in all sorts of ways... prescribing expensive drugs, employing more nursing staff, more dieticians, more counsellors, more dialysis... For those of us who work in an environment where money is an issue, if we are going to prescribe expensive drugs (in the UK the non-calcium binders are around ten times as expensive as calcium binders) we need to have solid data to show that these drugs have a significant impact on outcomes that matter to patients such as death, fractures, hospitalisations, quality of life.. I am afraid I see no robust data for the non-calcium binders
iv) It is entirely appropriate for pharmaceutical companies to push agendas that are going to maximise sales of their products. As Des Spence says we need experts who have a sense of scepticism and are not simply 'educational mercenaries'. But more than that as individual nephrologists we have a duty not to blindly accept expert opinion in the absence of data....
1. Effects of Phosphate Binders in Moderate CKD. Geoffrey A. Block, David C. Wheeler, Martha S. Persky, Bryan Kestenbaum, Markus Ketteler, David M. Spiegel, Matthew A. Allison, John Asplin, Gerard Smits, Andrew N. Hoofnagle, Laura Kooienga, Ravi Thadhani, Michael Mannstadt, Myles Wolf and Glenn M. Chertow. JASN online July 2012
Anyone who read my ASN blog will know that Remuzzi's group from Bergamo presented what appeared to be very positive data on the role of Rituximab in idiopathic membranous glomerulonephritis (IMN) at the ASN in 2011. Well the data has finally been published and serves as a salutory reminder to me that one always needs to see the final paper before being able to assess the impact on ones' practice. Remuzzi presents data on a 100 consecutive patient with IMN treated with Rituximab. The median duration of persistent proteinuria prior to rituximab therapy was over 2 years and follow up data was available for a median of 29 months. 27% achieved complete remission whilst 38% achieved partial remission (defined as less than 3g/24 hours and a 50% reduction in proteinuria from baseline). The authors suggest that 'Rituximab might be considered as first line therapy' for IMN. Does the data justify such a statement? Almost certainly not and its worth considering the following:
i) Given the very variable remission and progression rate with IMN, any therapeutic data without a control group is of limited value (some would say meaningless...). IMN runs a highly variable course with a high rate of spontaneous remission that occur can occur even quite late after presentation. This was observed not only by the GLOSEN group in Spain but by Remuzzi himself who when publishing in the NEJM on presumably another cohort of of 100 IMN patients found that 65% of patients went into complete or partial remission after 5 years of observation prompting him to note that 'Most untreated patients with idiopathic membranous nephropathy maintain renal function for prolonged periods and are likely to have spontaneous remission'......
ii) Following on from the above its worth noting that the serum creatinine and 24 hour urinary protein excretion was lower and the serum albumin was higher at baseline in the complete remission group versus the no remission group.. so the question arises did the complete remission group have milder disease that would just get better anyway. Of course unless you have a control you just dont know!
ii) Only 32 patients had recieved prior immunosuppression and of these only 24 had prior steroids in combination with an alkylating agent - this is currently the KDIGO recommended first line therapy as as Gerald Appel points out in an accompanying editorial, given the fact that we have placebo controlled RCTs with both CNIs and alkylating agents, but not with Rituximab, why should we use Rituximab as first line therapy?
iii) the authors cite safety concerns of cyclophosphamide although the cumulative dose of cyclophosphamide used in IMN is relatively low and the safety of Rituximab in auto-immune disease (in terms of side effects such as progressive multifocal leucoencepholopathy) maybe different than in those with lymphoma.
iv) Rituximab is expensive. The authors state that as they only use one dose (375mg/m2) costs are reduced though its worth noting that 18 patients with rituximab relapsed and required a second dose. The use of phospholipase A2 receptor antibodies as a guide to dosing schedule may make more sense.
Taking this work together with Fernando Fervenza's data from the Mayo Rituximab is probably better than placebo in IMN. Its expensive and until there is an RCT comparuing it with an alkylating agent or a CNI I would only use in those patients with progressive IMN who are refractory to (or have contraindications to) alkylating agents therapy and/or calcinuerin inhibitors.
1.Rituximab in Idiopathic Membranous Nephropathy. Piero Ruggenenti, Paolo Cravedi, Antonietta Chianca, Annalisa Perna, Barbara Ruggiero, Flavio Gaspari, Alessandro Rambaldi, Maddalena Marasà, and Giuseppe Remuzzi. J Am Soc Nephrol published 19 July 2012, 10.1681/ASN.2012020181
2. Prognosis of untreated patients with idiopathic membranous nephropathy. Schieppati A, Mosconi L, Perna A, Mecca G, Bertani T, Garattini S, Remuzzi G, N Engl J Med. 1993 Jul 8;329(2):85-9.
3. Spontaneous remission of nephrotic syndrome in membranous nephropathy with chronic renal impairment. Nephrol Dial Transplant 2012 Jan;27(1):231-4. Epub 2011 May 30. Polanco N, Gutiérrez E, Rivera F, Castellanos I, Baltar J, Lorenzo D, Praga M, Grupo de Estudio de las Enfermedades Glomerulares de la Sociedad Española de Nefrología (GLOSEN)
4. Rituximab therapy in idiopathic membranous nephropathy: a 2-year study. Fervenza FC, Abraham RS, Erickson SB, Irazabal MV, Eirin A, Specks U, Nachman PH, Bergstralh EJ, Leung N, Cosio FG, Hogan MC, Dillon JJ, Hickson LJ, Li X, Cattran DC; Mayo Nephrology Collaborative Group. . 2010 Dec;5(12):2188-98.
BLOG POSTED BY PROF ALAA SABRY:
This study by Giovanni Stallone1,etal published early online -NDT Advance Access published July 10, 2012 added to the 3 previously published studies about the role of Rapamycin for the treatment of ADPKD.
In this prospective, open-label randomized clinical trial, 55 ADPKD patients followed in the outpatient clinic of the two participating centres (University of Foggia and Bari, Italy) were enrolled between November 2007 and November 2008. The duration of the planned follow-up was
The inclusion criteria were clinical, genetic and ultrasonographic diagnosis of type I ADPKD, age between 18 and 65 years and an estimated glomerular filtration rate (eGFR) between 40 and 80 mL/min/1.73 m2, evaluated by the abbreviated Modification of Diet in Renal Disease (MDRD) formula
The primary objectives were to assess whether rapamycin may reduce the progressive increase in single cyst and total kidney volume in type I ADPKD and the decline in renal function and to identify the optimal rapamycin dose to achieve the beneficial results without incurring inmajor side effects.
After a screening assessment, all patients entered a run-in phase of 2 months in which eGFR and 24 h proteinuria were evaluated every 2 weeks. At the beginning of this period, in patients already treated with angiotensin-converting enzyme inhibitors (ACE-I) and angiotensin II receptor blockers (ARBs), these drugs were withdrawn for at least 4 weeks before performing the laboratory tests.
At the end of the run-in phase, all eligible patients were randomly assigned to ramipril alone (Group A) or ramipril plus high-dose rapamycin (Group or ramipril plus low-dose rapamycin (Group C). In Group B, rapamycin administration was given with a loading dose of 3 mg. The maintenance dose was 1 mg/day aiming at a blood trough level between 6 and 8 ng/mL. In Group C, the patients did not receive a
loading dose and the maintenance dose of rapamycin was 1 mg/day to reach a blood trough level between 2 and 4 ng/mL.
Magnetic resonance imaging technique .All patients recruited for our study underwent a standardized magnetic resonance imaging (MRI) examination at the baseline (T = 0) and after 24 months of initiation of drug therapy (T = 24) with the same MRI scanner.
DNA isolation and linkage analysis. Genomic DNA was derived from whole blood using Purelink Genomic DNA Mini kit (Invitrogen), according to the manufacturer’s protocol.
Epidermal growth factor urinary concentration:
was evaluated by ELISA and the results were normalized to urine creatinine excretion.
The mean rapamycin trough levels were 6.4 ± 0.3 ng/mL for Group B and 3.2 ± 0.4 ng/mL for Group C. The differences between the two groups were statistically significant at each time point for Group C.
In all three groups of patients, They observed an increased total kidney volume. Specifically,
in Groups A and B, there was a statistically significant increase in kidney volume after 24 months of treatment (P = 0.003 and 0.02, respectively), while in Group C, the increase failed to reach the statistical significance However, they did not observe any significant difference in change of kidney volume among the three study groups at the end of the observational period.
On the contrary, the cyst volume was increased in Group A patients after 24 months of treatment (P < 0.0001 versus basal), whereas it was significantly reduced in Group B (P < 0.0001 versus basal) and Group C (P <0.0001 versus basal) Also, in the case of single cyst volume, they did not observe any
significant difference among the three study groups at the end of the observational period.
Renal function analysis. They observed a decline in estimated creatinine clearance in Group A patients (P = 0.01, T24 versus T0),whereas in Groups B and C after 24 months, they observed
a slight increase in GFR, although the differences did not reach statistical significance .
The changes in the estimated creatinine clearance in the three groups at
the end were not significantly different.
EGF urine excretion. an increased EGF urinary excretion in Group A patients (P =0.001 versus T0) was observed, whereas the treatment groups were characterized by a significant reduction after 24 months
(Group B: P = 0.0001 versus T0; Group C: P = 0.0001 versus T0)
The results of this pilot study suggest that rapamycin treatment at a dose effectively inhibiting p70S6 kinase in circulating PBMCs does not significantly slow down cyst growth and renal function decline featuring the natural history of ADPKD
Both chronic kidney disease (CKD) and osteoporosis are emerging as major public health problems associated with an ageing population. Osteoporosis is characterised by reduced bone strength resulting in an increased predisposition to fragility fracture. Bone strength itself is a reflection of both bone mineral density (BMD) and bone quality. There is significant co-prevalence of osteoporotic fractures and CKD in the elderly population. The qualitative abnormalities in bone in CKD are reflected in the fact that fracture risk is much higher in both the dialysis and the non-dialysis CKD population and the outcomes of fractures are significantly worse with a 2 to 3 fold increase in mortality after hip fracture in the CKD population compared to the general population.
However the management of bone disease in CKD remains uncertain - the fundamental problem being that we simply do not know whether achieving biochemical 'targets' (say PTH, calcium phosphorus) or whether the use of therapeutic agents (such as bisphosphonates, calcimimetics or vitamin D) have any impact on outcomes that matter to patients - like death or fracture rates. Furthermore apart from bone biopsy (which is rarely performed) we don't have any tools to assess bone quality and predict fracture risk.
A recent paper from Yenchek and colleagues is published in this months cJASN and is the first prospective, longitudinal study to specifically test the hypothesis that bone mineral density (BMD) predicts fracture risk irrespective of CKD status. In a cohort of 2754 (of whom 587 had CKD stages 3-5) patients aged between 70-79 years old data was collected on incident fractures over a follow up of 11 years. There were 384 incident fractures and even after adjustment for factors such as age, sex, race, BMI, vitamin D and PTH, low femoral neck BMD as measured by DXA (Dual Energy X-Ray Absorptiometry)was associated with fracture risk irrespective of CKD status. The authors state "This argues against the current KDIGO guideline recommendations and suggests that there may be a role for DXA screening in CKD."
So should we start using DXA to identify fracture risk in CKD and prescribe anti-resorptive therapy to those with reduced BMD? Well I'm not sure I share their view and it is worth considering the following:
i) The study by Yenchek and colleagues is important but only 4% of patients had CKD4/5 so the jury remains out on whether DXA will be useful in CKD4/5
ii) We know from post-hoc analysis from the large registration studies that both Bisphosphonates and Denosumab can reduce fracture risk in osteoporotic patients with CKD3 and CKD4 (down to an eGFR of around 20mls/minute). However, it is worth noting the all patients with abnormal PTH or vitamin D levels were excluded from these studies. Therefore whilst these drugs probably reduce fracture risk in patients with CKD3/4 with 'pure' osteoporosis we simply do not know whether they have any impact in those with evidence of coexistent renal osteodystrophy.
iii) Whilst DXA may identify fracture risk in CKD3 a careful cost/benefit analysis study would need to be performed before such screening could be recommended as routine - indeed it may be cheaper just to focus on non-pharmacological strategies such as reducing risk of falls and improving neuromuscular strength.
iv) A key challenge is how to develop the evidence base for reducing fracture rates in CKD 4 and 5. In the general population huge studies with thousands of patients were required to show a beneficial effect of anti-resorptive therapy. The reality is that it will be almost impossible) to carry out such large studies in the CKD4/5 population - therefore perhaps a more realistic goal is to develop validated surrogate markers of fracture risk in this population (be they biochemical markers of bone turnover combine with bone biopsy data or imaging such as high resolution CT) and then evaluate whether modifying these factors can impact of fracture rates.
- Yenchek RH, Ix JH, Shlipack MG, Bauer DC, Rianon NJ, Kritchevsky SB, Harris TB, Newman AB, Cauley JA, Fried LF, for the Health, Aging, and Body Composition Study: Bone mineral density and fracture risk in older individuals with CKD. Clin J Am Soc Nephrol 7: 1130–1136, 2012
- Miller PD, Roux C, Boonen S, Barton IP, Dunlap LE, Burgio DE: Safety and efficacy of risedronate in patients with age-related reduced renal function as estimated by the Cockcroft and Gault method: a pooled analysis of nine clinical trials. J Bone Miner Res 20: 2105–2115, 2005
- Jamal SA, Bauer DC, Ensrud KE, Cauley JA, Hochberg M, Ishani A, Cummings SR: Alendronate treatment in women with normal to severely impaired renal function: an analysis of the fracture intervention trial. J Bone Miner Res 22: 503–508, 2007
- Jamal SA, Ljunggren O, Stehman-Breen C, Cummings SR, McClung MR, Goemaere S, Ebeling PR, Franek E, Yang YC, Egbuna OI, Boonen S, Miller PD. J Bone Miner Res. 2011 Aug;26 (8): 1829-35
In the June issue of cJASN Rahman and colleagues from the ALLHAT Collaborative Group report the long-term (9 year) renal and cardiovascular outcomes of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT).
This is a randomised, double blind, multicentre clinical trial that compares the impact of 5 year treatment with an ACE inhibitor (lisinopril), a calcium antagonist (Amlodipine), an alpha blocker (Doxazosin) and a thiazide diuretic (Chlorthalidone) on the incidence of major coronary heart disease (CHD) in high risk hypertensive patients. The initial analysis over a 5 years observation and follow-up period showed that neither lisinopril not amlodipine were superior to the diuretic chlorthalidone. The Doxazosin arm had to be discontinued earlier due to side effects and increased mortality (ALLHAT Collaborative Research Group: JAMA 283:1967-1975, 2000).
The current posthoc analysis reports on 31,350 participants followed up for up to 9years, based on hospitalisation records, in term of cardiovascular outcomes; total mortality, CHD, CVD, Stroke, Heart Failure and ESRD.
In the whole study group, there was no difference between lisinopril, amlodipine and chlorthalidone isn relation to CVD or renal outcomes/ESRD. Subgroup analysis of pateints with CKD and an eGFR<60 (with minimal proteinuria), also showed no difference in renal or cardiovascular outcomes between the groups. Of interest, the morbidity and mortality from CVD causes was similar to those observed in the RENAAL and AASK trials.
The authors conclude that independently of BP lowering effect, no antihypertensive drug class has significant advantage over others in preventing CVD; morbidity and mortality. Of note amlodipine was less effective than chlorthalidone in preventing heart failure; although this may simply reflect that the former favours fluid retention whilst the latter minimises it...!!!
This observation agrees with a review of controlled trials by Segall et al (J Nephrol 2008;21:374-383) that showed no specific benefit of anti-hypertensive drug class beyond BP control. Other studies suggested a therapeutic advantage of RAAS inhibition on CVD but it was unclear whether this was independent from improved BP control. Of note the much quoted HOPE study, also failed to dissociate the CVD protective effect of Ramipril from its antihypertensive impact. More recently, HOPE study posthoc analysis based on 24h ambulatory BP Monitoring failed to convincingly dissociate cardioprotection from improved BP control (Svensson et al. Hypertension. 2001;38:e28-e32).
From the above one is entitled to ask whether the much publicised cardioprotective effect of RAAS inhibitors is real or a bais generated by:
1. Poorly designed clinical trials (RCT); with unequal BP control between groups
2. Inadequate BP measurements in RCT; with occasional office BP measurements now known to be poorly reflective of 24h BP recordings measuring nocturnal BP and nocturnal dip in BP, both known to be closely associated with CV events.
3. An industry led brainwashing of ill informed Nephrologists; who follow the trend without asking questions....?!
I suspect the answer lies in all three...!!!